Policy Research Working Paper 5875 Should Cash Transfers Be Confined to the Poor? Implications for Poverty and Inequality in Latin America Pablo Acosta Phillipe Leite Jamele Rigolini e World Bank Latin America and Caribbean Region Office of the Chief Economist November 2011 WPS5875 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized
35
Embed
Should Cash Transfers Be Confined to the Poor?...Bolsa Familia Poor and extreme poor households 30 per household + 7 per child 11,100,000 households 0.36. Chile. Chile Solidario Extreme
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Policy Research Working Paper 5875
Should Cash Transfers Be Confined to the Poor?
Implications for Poverty and Inequality in Latin America
Pablo AcostaPhillipe Leite
Jamele Rigolini
The World BankLatin America and Caribbean RegionOffice of the Chief EconomistNovember 2011
WPS5875P
ublic
Dis
clos
ure
Aut
horiz
edP
ublic
Dis
clos
ure
Aut
horiz
edP
ublic
Dis
clos
ure
Aut
horiz
edP
ublic
Dis
clos
ure
Aut
horiz
ed
Produced by the Research Support Team
Abstract
The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.
Policy Research Working Paper 5875
This paper compares for 13 Latin American countries the poverty and inequality impacts of cash transfer programs that are given to all children and the elderly (that is, “categorical” transfers), to programs of equal budget that are confined to the poor within each population group (that is, “poverty targeted” transfers). The analysis finds that both the incidence of poverty and the depth of the poverty gap are important factors affecting the relative effectiveness of categorical versus poverty targeted transfers. The comparison of transfers to children and the elderly also supports the view that
This paper is a product of the Office of the Chief Economist, Latin America and Caribbean Region. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be contacted at [email protected].
choosing carefully categories of beneficiaries is almost as important as targeting the poor for achieving a high poverty and inequality impact. Overall, the findings suggest that although in the Latin American context poverty targeting tends to deliver higher poverty impacts, there are circumstances under which categorical targeting confined to geographical regions (sometimes called “geographic targeting”) may be a valid option to consider. This is particularly the case in low-income countries with widespread pockets of poverty.
Policy Research Working Paper 5875
Should Cash Transfers Be Confined to the Poor?
Implications for Poverty and Inequality in Latin America
Pablo AcostaPhillipe Leite
Jamele Rigolini
The World BankLatin America and Caribbean RegionOffice of the Chief EconomistNovember 2011
Produced by the Research Support Team
Abstract
The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.
Policy Research Working Paper 5875
This paper compares for 13 Latin American countries the poverty and inequality impacts of cash transfer programs that are given to all children and the elderly (that is, “categorical” transfers), to programs of equal budget that are confined to the poor within each population group (that is, “poverty targeted” transfers). The analysis finds that both the incidence of poverty and the depth of the poverty gap are important factors affecting the relative effectiveness of categorical versus poverty targeted transfers. The comparison of transfers to children and the elderly also supports the view that
This paper is a product of the Office of the Chief Economist, Latin America and Caribbean Region. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be contacted at [email protected].
choosing carefully categories of beneficiaries is almost as important as targeting the poor for achieving a high poverty and inequality impact. Overall, the findings suggest that although in the Latin American context poverty targeting tends to deliver higher poverty impacts, there are circumstances under which categorical targeting confined to geographical regions (sometimes called “geographic targeting”) may be a valid option to consider. This is particularly the case in low-income countries with widespread pockets of poverty.
<okay – sg>
SHOULD CASH TRANSFERS BE CONFINED TO THE POOR?
IMPLICATIONS FOR POVERTY AND INEQUALITY IN LATIN AMERICA1
Pablo Acosta Phillipe Leite Jamele Rigolini
The World Bank
1 We would like to dedicate this paper to Gozalo Llorente, who unexpectedly left us during its writing. We
would like to thank Gonzalo and Alinne Veiga for excellent research assistance. We also thank helpful comments from Margaret Grosh, Ruslan Yemtsov, and Javier Baez. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.
2
1. Introduction
Cash transfer programs have become the centerpiece of many Latin American countries’
social protection agenda. They have become popular not only as short term instruments to
help the poor cope with economic shocks, but also as longer-term poverty alleviation
programs supporting minimum consumption levels and promoting the accumulation of
human capital.
Yet, while several evaluations have demonstrated the impact of cash transfers on
poverty reduction and human development outcomes, they do absorb an important share
of governments’ budgets (Grosh et al., 2008; Fiszbein and Schady, 2010; World Bank-IEG,
2011). Most countries in the Latin America and Caribbean (LAC) region count with
conditional cash transfer programs (CCTs) targeted to poor households with children that
account for budgets up to 0.6 percent of GDP (see Table 1). Some countries in the region
(fewer than those with CCTs) also provide cash transfers to the elderly, many of them on a
universal (i.e. categorical) basis, that require an even more sizeable share of public budget –
up to 1.3 percent of GDP (see Table 2). These programs can account, in a country like Brazil,
for up to 1.7 percent of GDP, and for the average in the region, between 1 and 2 percent of
GDP (Grosh et al., 2008).
Resource constraints and ethical considerations have led to a heated debate on the
scope and coverage of cash transfers. Few doubt about providing benefits to people falling
within demonstrable categories of vulnerable groups, such as the children, the elderly or the
disabled, though there is still an open debate on how children and the elderly compete as
the primary group to focus social assistance.2 But there is less consensus on whether
transfers should be given to all people within these categories (i.e. be “categorical”), or if
they should be restricted to poor people within categories of beneficiaries (i.e. be “poverty
targeted”). While poverty targeted transfers are more cost effective because limited
resources are distributed among fewer beneficiaries, they remain more complex and costly
to administer. And even the most sophisticated targeting systems miss some of the poor,
2 Some argue that political lobbying tends to favor the elderly (“the elderly can vote, the children cannot:”
Preston, 1984). Even in countries with well-developed social assistance schemes with a family accompaniment approach like Brazil and Chile, public spending per older is significantly higher than public spending per child (Turra et al., 2011).
3
and may have difficulties to adapt to entry and exit to and from poverty because registers of
beneficiaries cannot be updated frequently.
Table 1: Conditional Cash Transfers in Latin America and the Caribbean
Source: Based on Fiszbein and Schady (2010).
It is not therefore always obvious that poverty-targeted social assistance programs are the
best approach to alleviate poverty. At the heart, the optimal design relates to societal
preferences for redistribution and taxation (Mkandawire, 2005), as well as tolerance to
exclusion errors or deviating from a “rights” approach. But the choice of design can also be
informed by technical considerations, such as looking at the accuracy and cost effectiveness
of different targeting mechanisms, which is the focus of this paper.
Country Program Target PopulationAmount
(monthly US$)
Number of Beneficiaries
(latest available)
Cost
(% of GDP)
Argentina Programa FamiliasHousehold Heads, Pregant
Females, Children <1940-80 per child 500,000 households
Bolivia Juancito PintoPublic school children up to
grade 62 per child 500,000 households
Brazil Bolsa FamiliaPoor and extreme poor
households
30 per household +
7 per child11,100,000 households 0.36
Chile Chile SolidarioExtreme poor households
(means-tested)14 per child 256,000 households 0.08
Colombia Familias en AccionExtreme poor households
with children <7 (health), and
with children <17 (education)
Education: 8-33 per
child; Health: 28 per
household
1,700,000 households 0.20
Dominican
RepublicSolidaridad
Poor and extreme poor
households with children <1729 per household 461,000 households
EcuadorBono de Desarrollo
Humano
Households in first 2 income
quintiles with children <1715 per household 1,060,000 households 0.60
El SalvadorComunidades
Solidarias Rurales
Extreme poor households
with children <16 in 100 rural
municipalities
Education: 15 per
household; Health:
15 per household
100,000 households
Guatemala Mi Familia ProgresaExtreme poor households
with children <16 in 130
municipalities
Education: 20 per
household; Health:
20 per household
250,000 households 0.20
HondurasPrograma de
Asignacion Familiar
Poor households with
children 6-12 years old up tp
grade 4
Education: 5 per
household; Health:
4 per household
240,000 households
JamaicaProgram of
Advancement through
Health and Education
Poor households (means-
tested) until they graduate
from secondary school
100 per child 100,000 households
Mexico OportunidadesExtreme poor households
(means-tested)
Education: 12-23
per household;
Health: 17 per
5,000,000 households 0.40
PanamaRed de
Oportunidades
Extreme poor households
(means-tested)50 per household 70,000 households
Paraguay PROPAIS IIExtreme poor households
with children <15 in rural
areas
120 per household 5,800 households 0.08
Peru JuntosPoor households with
children <1533 per household 454.000 households 0.11
4
Table 2: Non-Contributory Social Pensions in Latin America and the Caribbean
Source: Murrugarra (2011).
This paper simulates impacts of categorical and poverty targeted cash transfers on poverty
and inequality in 13 Latin American countries: Argentina, Bolivia, Brazil, Chile, Colombia,
Costa Rica, the Dominican Republic, Ecuador, Guatemala, Mexico, Nicaragua, Panama, and
Peru. The simulations focus on two programs that tend to be the most widespread in the
region: transfers to children up to five years of age, and to elderly people that are older than
65.
While, by their nature, poverty targeted transfers always deliver a higher poverty
impact, both the incidence of poverty and the depth of the poverty gap appear to be
important factors affecting the relative effectiveness of categorical vs. poverty targeted
transfers. The comparison of transfers to children and the elderly also supports the view
that choosing carefully categories of beneficiaries is almost as important as targeting the
poor for achieving a high poverty impact. Overall, the findings suggest that in the Latin
Country Program Target PopulationTransfer
(monthly US$)
Number of Beneficiaries
(latest available)
Cost
(% of GDP)
Argentina Pensiones No
Contributivas
Asistenciales por
65+ without contributory
pension and in poverty
151.5 65,900
Bolivia Renta Dignidad
(former Bonosol)
60+, universal 22-29 782,660 1.30
Brazil Rural Pension 65+ in rural areas 342 800,000
Chile Pension Basica
Solidaria
65+, in 3 lowest quintiles of
income distribution
150 407,000 0.50
Costa Rica Regimen No
Contributivo
65+ in poverty 135 53,492 0.24
Ecuador Pension Asistencial 65+ without contributory
pension
35 502,828
El Salvador Pension Basica
Universal
70+ in severe extreme poor
municipalities
50 19,534
70 y mas 70+ in selected municipalities,
universal
28.5 2,000,000
Oportunidades Adulto
Mayor
70+ in households receiving
CCTs (Oportunidades)
22.7 80,000
Pension Alimentaria 70+ in Mexico DF, universal 63.2 470,000
Peru Gratitud 75+ without contributory
pension and in poverty
36 0.03
Trinidad and
Tobago
Senior Citizen Pension 65+, income means-tested 189-472 73,110 1.30
Uruguay Beneficio No
Contributivo
70+, Income means-tested 240 31,577
Mexico
5
American context targeting assistance to the poor tends to deliver higher poverty impacts.
There are nonetheless circumstances under which categorical targeting confined to
geographical regions may be a valid option to consider. This is particularly the case in low
income countries with widespread pockets of poverty.
The paper proceeds as follows. The next section summarizes trends in social assistance
and the ongoing debate on categorical vs. targeted social transfers. Section 3 discusses the
methodology used in the simulations. Section 4 presents the simulation results for the 13
Latin American countries. Section 5 concludes.
2. The cash and targeting revolutions
All over the developing world, and particularly in Latin America, social protection programs
are moving from universal in-kind benefits and subsidies, to targeted cash transfers. The
rationale behind the silent revolutions towards cash lies in higher welfare impacts (cash can
support households’ needs beyond food), economic efficiency (e.g., avoidance of dead-
weight losses and distortions from subsidies), transparency and accountability, and
simplification of the administrative procedures in service delivery (Grosh et al., 2008). While
cash transfers are far from being a panacea and should be designed in the context of a
broader social assistance strategy that takes into account the local context and political
economy, they undoubtedly improved in many countries both beneficiaries’ welfare and the
cost effectiveness of social assistance programs.
With the expansion of cash transfers, a debate has however arisen on the extent to
which assistance should be provided to all people falling within demonstrable vulnerable
categories (such as the children, the elderly or the disabled), or only to the poor within
these categories. Is it optimal, and, equally important, is it ethically correct to handpick
beneficiaries within categories?
From an ethical perspective, many advocate that social assistance programs should
favor horizontal equity and that any person falling into a category that tends to be
vulnerable should have the right to receive assistance (ILO-UN Social Protection Floor
initiative, 2010). This universal view of social assistance is strengthened by three potential
6
drawbacks of confining assistance only to poor beneficiaries within each category. First,
poverty targeting may generate behavioral distortions and induce informality. For instance,
poverty targeted pensions and health insurance programs financed out of general taxation
that coexist with contributory ones may generate incentives to remain in the informal labor
market and thus avoid contributing to the system (Levy, 2008). Poverty targeted cash
transfers may also discourage labor force participation. Second, there is no targeting system
that can perfectly identify vulnerable individuals within a given category. Even the most
sophisticated targeting mechanisms miss some of the poor (exclusion error), and include
some wealthier individuals (inclusion error). The question there is up to which point a
society is ready to exclude some of the poor from assistance because of efficiency
considerations. Finally, potential stigmas related to poverty targeting may also affect
participation of the most vulnerable (Grosh et al., 2008). Leite (2011) finds, for instance,
that in Tanzania some of the elderly from rural poor villages did declare having 3 meals a
day to avoid the stigma of being perceived as poor.
The extent to which these drawbacks represent a strong argument against targeting still
remains an open discussion. When the benefits of social assistance programs do not
become disproportionate, existing studies find that labor market distortions from poverty
targeting remain moderate (Parker and Skoufias, 2000; Skoufias and Di Maro, 2011; Fiszbein
and Schady, 2010; World Bank-IEG, 2011). And while exclusion error may still be
considerable, many categorical programs also miss some of the poor – though in lower
proportions – because of information failures and high participation costs in remote areas
(Barrientos, 2008; Sluchynsky, 2008).
The categorical approach is also challenged by the reality of hard budget constraints and
political economy considerations. Countries have limited resources to fight poverty and
promote equity. Broad social assistance policies in developing countries cost already on
average 2 percent of GDP (Weigand and Grosh, 2008; Grosh et al., 2008), though wide
disparities subsist across countries.3 And even if some countries may have the fiscal space to
expand further coverage, this can be politically costly. In general, policy discussions tend to
center on how to improve the programs’ impact within the existing resources.
3 Programs focused on the elderly poor range for instance from 0.1 percent of GDP in the Seychelles to 10.6
percent of GDP in Ethiopia. See Schwarz (2003), and Kakwani and Subbarao (2005).
7
In addition to ethical and political economy considerations, there are also technical
arguments both in favor and against targeting. The main advantage of poverty targeting is
to increase the amount that can be transferred to each beneficiary for a given budget, which
maximizes the poverty impact of transfers. For instance, Grosh and Leite (2008) analyze
cash transfer for the elderly in four countries – Yemen, Niger, Panama and Kyrgyzstan. They
find that, despite exclusion errors, poverty targeted social pensions are much more cost-
effective per dollar spent, and, for a fixed budget, convey a higher poverty impact. On the
other hand, however, targeting is costly. Caldes et al. (2006) and Grosh et al. (2008) find
that administrative costs of poverty targeted conditional cash transfers (including the cost
of collecting the data needed to construct a proxy means test and periodical reclassification
of beneficiaries) tend to be around 10 percent of the program’s budget. In some extreme
cases it can absorb up to 30 percent of it, which can seriously affect the ability to provide